Identification of HLA-A*0201-restricted CTL epitopes from the receptor-binding domain of MERS-CoV spike protein using a combinatorial in silico approach

Abstract

A novel SARS-like illness called Middle East respiratory syndrome coronavirus (MERS-CoV), caused by an emerging coronavirus, has been a recent cause for concern due to its fatality and pandemic potential. Developing a peptide-based vaccine could be helpful in fighting against the virus. Since the experimental procedure is time-consuming and expensive, computational analysis can play an important role in accelerating the process. Therefore, the aim of this study was to computationally identify cytotoxic T-lymphocyte epitopes presented by the human leukocyte antigen (HLA)-A*0201, as this is the most frequent HLA class I allele among Middle Eastern populations. The receptor-binding domain of the spike glycoprotein of MERS-CoV, by which the virus binds to its entry receptor to further infect host cells, is a potential candidate used here for running our in silico epitope identification process. The results include predicted epitopes together with their interaction properties with major histocompatibility complex (MHC) molecules and also the binding behavior of MHC-epitope complexes to human T-cell receptor. Predicted epitopes with the most preferable binding properties are beneficial for vaccine development. Therefore, the huge experimental workload for epitope-based vaccine design will be minimized.

A Computational Vaccine Designing Approach for MERS-CoV Infections

“…Prediction of cytotoxic T-lymphocyte epitopes and their interaction with MHC Class I, the results showed ILDYFSYPL was similar according my study, Badwai et al [16] and Poorinmohammad and Mohabatkar [15] studies; partially similarity with Iranian study [15] in LLSGTPPQV, ILATVPHNL, LQMGFGITV, and FSNPTCLIL epitopes were noticed except NLTTITKPL epitope that was absent from my study in S and modified S sequence; FSNPTCLIL represents the only epitope that is found in my study in S and modified S sequence; FSFGVTQEY have a high affinity to bind to many alleles and these findings agree with Badawi et al [16] in addition to ITYQGLFPY in my study through S glycoprotein sequence, but still there are differences in the numbers of selected epitopes that reacted with MHC-I which were higher than that in Badawi et al [16], while in E protein FIFTVVCAI epitope has a higher allele affinity followed by ITLLVCMAF, IVNFFIFTV, and LVQPALYLY reverse modified E protein; LVQPALSLY epitope has shown high affinity and then followed by LYMTGRSVY, WFIPNFFDF, YMTGRSVYV, ITLLVCTAF, FVQERIGWF, FLTATHLCV, and CMTGFNTLL, the last epitope which is common between E and modified E protein sequences.…”

“…Today, there are so many different ways to develop MERS-CoV vaccine; some of them partially succeed but the others failed while the remaining nor succeed neither failed because it depends on software program for different reasons and still need to go under vaccine protocols processing, in those studies that consist with S1 protein subunit especially RBD (the most mutable region that containing mutation sites which define antibody escape variants) was considered the basis for several MERS-CoV vaccine candidates in many studies such as using RBD with aluminum salt or oil-inwater adjuvants; can elicited neutralizing antibodies of high potency across multiple viral strains by Modjarrad [4] and Wang et al [6] said that the full-length S DNA and a truncated S1 subunit glycoprotein can elicit a higher titer of neutralizing antibodies; this kind of immunization protected non-human primates (NHPs) from severe lung disease after intratracheal challenge with MERS-CoV injection; in another study that was done in Iran by Poorinmohammad et al [15] [NetCTL 1.2 (Larsen et al, 2007), EpiJen (Doytchinova et al, 2006), and NHLApred (Bhasin and Raghava, 2007), they were selected computational prediction tools with PEPstr server for modeling (Kaur et al, 2007)] to identify cytotoxic T-lymphocyte epitopes presented by the human leukocyte antigen (HLA)-A * 0201; as this is the most frequent HLA class I allele among Middle Eastern populations with this selected RBD for their study, they showed LLSGTPPQV, ILDYFSYPL ILATVPHNL, NLTTITKPL, LQMGFGITV, and FSNPTCLIL as selected epitopes but LLSGTPPQV and FSNPTCLIL were considered as real epitope due to the following: peptides with binding orientations closer to the native structure and lower binding free energy scores are ranked higher in having the potential to be real epitopes reverse another study were done by Shi J et al [19] by using the Immune Epitope Database, that said: the nucleocapsid (N) protein of MERS-CoV might be a better protective immunogen with high conservancy and potential eliciting both neutralizing antibodies and T-cell responses when compared with spike (S) protein; in addition 71 peptides were identified as helper T-cell epitopes, 34 peptides were identified as CTL epitopes; just top 10 helper T-cell epitopes and CTL epitopes based on maximum HLA binding alleles, can elicit protective cellular immune responses against MERS-CoV were considered as MERS vaccine candidates and they are covering 15 geographic regions [19]. In this study that consists of two parts reference and modified sequence of both S glycoprotein and E protein, I found that the most common B-cell epitope that passed all B-cell prediction methods [IEDB prediction tool] for E protein is YVKFQDS in position 69 and for modified E they are VYVPQQD, YVPQQDS, and PPLPED/PPLPEDV epitopes at positions 68, 69, and 77 sequentially; while for S and modified S, they are DVGPDSV, PDSVKSA, DSVKSAC, PRPIDVS, HTPATDC, AKPSGSV, KPSGSVV, SGTPPQV, GTPPQVY, TPPQVYN, QLSPLEG, YGPLQTP, PRSVRSV, RSVRSVP, SVKSSQS, VKSSQSS, SQSSPII, and SLNTKYV at positions 23,26,27,48,211,371,372,…”

“…The positive result number of selected peptide represents 668 out of 1348 in S glycoprotein, while it represents 19 out of 76 in E protein and 673 out of 673 with 21 out of 76 in both S and E modified sequence sequentially. 773 are ordinary only found in S glycoprotein, while LTPTSSY, TPTSSYV, PTSSYVD, TSSYVDV, DHGDYYV, YSQDVKQ, ANQYSPC, NQYSPCV, and YYRKQLS in a positions 15,16,17,18,83,108,523,524, and 543 sequentially are only found in S glycoprotein modified sequence.…”

“…Spike glycoprotein, E protein, and modified S and E protein were subjected to BepiPred linear epitope prediction, Emini surface accessibility, Kolaskar and Tongaonkar antigenicity, Parker hydrophobicity, Chou and Fasman beta turn prediction methods, and Karplus and Schulz flexibility in IEDB, as the results in Figs. 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23, and 24.…”

Kremen is beyond a subsidiary co-receptor of Wnt signaling: an in silico validation

“…In conclusion, using in silico tools would pave the way to derive compelling results and hypotheses in various fields of biology (Ali et al, 2014;Liangbing et al, 2014;Poorinmohammad and Mohabatkar, 2014). Kremen receptors in tumor cells may determine either the antitumor or the protumor effect of the DKK protein, depending on which domains are involved.…”